9 Summary and Future Work

We have completed an optical identification program for FIRST radio sources by matching the FIRST catalog to a catalog constructed from the APM scans of the POSS-I survey. A detailed analysis of the properties of the APM catalog has been undertaken; we have recalibrated its photometry using the APS, and have produced a factor of $\sim10$ improvement in astrometric accuracy by using the FIRST sources as a set of astrometric standards. We have discussed in some detail the procedures for identifying radio source counterparts through positional coincidence, calculating false match rates for various subsets of both catalogs. For stellar counterparts of isolated radio sources, the matches are 90% reliable and 90% complete within an offset of $1.2\arcsec$ ; over 98% of all matches within $1\arcsec$ are correct identifications. In total, nearly 73,000 radio source identifications are recorded, an increase by a factor of $\sim200$ over the total number of counterparts previously reported for radio-selected samples in the flux density range 1-30 mJy. We present color-magnitude diagrams for the radio counterparts and briefly discuss the various source populations represented.

A catalog of all 454,754 optical objects within $20\arcsec$ of the 382,892 radio sources in the north Galactic cap region FIRST has covered to date is available on the FIRST Webpage. As we complete radio observations and optical matching for additional regions of the sky, this catalog will be expanded accordingly. The results presented here suggest that future work to match FIRST with the GSC-II and the Sloan Digital Sky Survey (with which the FIRST sky coverage is matched) will yield enormous, valuable samples of optically identified radio sources for use in a wide variety of applications from stellar radio astronomy to cosmology.

Acknowledgments

The success of the FIRST survey is in large measure due to the generous support of a number of organizations. In particular, we acknowledge support from the NRAO, the NSF (grants AST-98-02791 and AST-98-02732), the Institute of Geophysics and Planetary Physics (operated under the auspices of the U. S. Department of Energy by Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48), the Space Telescope Science Institute, NATO, the National Geographic Society (grant NGS No. 5393-094), Columbia University, and Sun Microsystems. We thank Juan Cabanela for his assistance in matching with the APS catalog. RGM acknowledges support from the Royal Society. DJH is grateful for the support of the Raymond and Beverly Sackler Fund, and joins RHB and RLW in thanking the Institute of Astronomy of the University of Cambridge for hospitality during much of this work. We thanks Mike Irwin for his explanatory comments on a draft of this paper. This paper is Contribution Number 685 of the Columbia Astrophysics Laboratory.